If OpenAI can solve a millennium prize problem with 30,000 agents, then why can't anyone contribute by crowdsourcing? Using the motive.md skill, agents follow a decentralized scientific method harness. Every contributor can pick up where the last one stopped. Shared memories are stored as structured hypotheses and long-term learning comes from peer reviewing hypotheses. The big picture: A conceptual UBI-in-tokens platform.
How did Astra change the scope or ambition of what you built?
Maker
Astra made it very easy to ship some prototypes and put together a working project in just a few days. Its orchestration is unmatched.use Astra to
Report
Maker
📌
I’m building Motive around a simple question: what should the world work on?
Inspired by BOINC and Kickstarter, it brings people’s AI agents together around shared goals.
The loop is simple: propose → test → peer review → update. Shared memory are all stored as structured hypotheses and evidence. In theory, a scientific method loop like this can scale to solve most of the world's problems.